T OXICOLOGIA GENETICA
R. Rita Misra y Michael P. Waalkes
Figure 23. Helmet Ownership Among Those That Do Not Always Wear Helmets
Respondents that did not always wear a helmet were further asked to define the main reason why they did not. Respondents were asked: “What is the MAIN REASON you do not always use a helmet while using bikesharing? Select the circumstances that most often apply to you regarding helmet use.” Respondents were given four options, which covered the general responses of: “I never wear helmet;” “My bikesharing is not always planned…;” “I do not like carrying a helmet;” and “Other, please explain.” The breakdown of responses is shown in Table 28.
Mineta Transportation Institute
86 Public Bikesharing User Survey Analysis
Table 28. Main Reason Why Respondents Do Not Wear a Helmet While Bikesharing
What is the MAIN REASON you do not always use a helmet while using bikesharing? Select the circumstances that most often apply to you regarding helmet use.
Response Options Montreal Toronto
Minneapolis-
Saint Paul Salt Lake City Mexico City
I never wear a helmet while riding any bicycle.
23% 14% 13% 2% 29%
My use of bikesharing is not always planned and I do not have a helmet with me in such cases.
42% 47% 45% 53% 27%
I do not like to carry a helmet around, even though I generally know in advance when I am going to use bikesharing.
29% 32% 31% 26% 25%
Other, please explain: 5% 6% 9% 19% 6%
I do not own a helmet. 2% 1% 2% 0% 13%
Total 893 792 506 43 3145
Within the “Other” response, a common write-in response was: “I do not own a helmet.” These responses were identified as a fifth category within each survey and are listed separately. Respondents in four of five surveys indicated that the most common reason for not always wearing a helmet was due to the unplanned nature of bikesharing trips. The second most common response was that respondents did not like carrying helmets around. The categorical distribution in Mexico City was slightly different, with the top response: “I never wear a helmet,” followed by “unplanned use” and “do not like carrying a helmet.” Notably, 13% of respondents wrote that the lack helmet ownership was a key inhibitor to using one, whereas far fewer cited this in the other cities.
For those respondents that simply answered: “I never wear a helmet while riding any bicycle,” the survey probed even further to understand why. Respondents were asked to rank the
top three reasons they never wore a helmet. The two most common responses that ranked number “1” were: “I am a very safe bicycle rider,” and “it is not necessary” and “I should probably get a helmet, but haven’t found the time to find one I like.” Overall, most responses indicated that people who never wear helmets, do so more by choice rather than constraint. Other responses available included: “helmets are uncomfortable,” “helmets mess up my hair,” and “helmets do not look good on me.” When aggregated together, these “choice- based” responses comprised over 60% of the selections by respondents in the U.S. and Canada, and 45% of respondents in Mexico City. Responses based on helmet availability encompassed between 15% to 30% of responses in the U.S. and Canada and 42% in Mexico. The responses are summarized in Table 29.
The most common reasons for not always wearing a helmet were the unplanned nature of tripmaking and that users did not like to carry a helmet around.
Mineta Transportation Institute
87 Public Bikesharing User Survey Analysis
Table 29. Ranked Reasons Why Respondents Never Wear a Helmet
Which of the following best describes why you do not wear a bicycle helmet?
Please rank the top three reasons, with one being the most important and three the least of your top choices.
Rank Montreal Toronto Minneapolis-Saint Paul Salt Lake City Mexico City
I am a very safe bicycle rider, and it is not necessary.
1 27% 37% 35% 0% 22%
2 17% 21% 20% 100% 20%
3 15% 14% 9% 0% 14%
Bicycle helmets are uncomfortable.
1 17% 13% 11% 0% 14%
2 19% 23% 26% 0% 21%
3 19% 13% 12% 0% 19%
Bicycle helmets mess up my hair when I wear them.
1 9% 15% 8% 100% 6%
2 10% 14% 18% 0% 11%
3 18% 14% 18% 0% 22%
Bicycle helmets do not look good on me.
1 8% 1% 6% 0% 2%
2 11% 7% 5% 0% 4%
3 18% 10% 5% 100% 15%
I cannot afford a bicycle helmet.
1 5% 0% 3% 0% 6%
2 7% 7% 0% 0% 9%
3 11% 0% 3% 0% 13%
I should probably get a helmet, but I haven’t found the time to find one I like.
1 22% 15% 22% 0% 36% 2 18% 8% 9% 0% 17% 3 16% 12% 17% 0% 14% Other 1 12% 19% 15% 0% 12% 2 4% 7% 12% 0% 5% 3 7% 9% 14% 0% 8% N 204 107 65 1 868
On-Street Intercept Survey Results
The authors developed an on-street survey experimental survey in an attempt to better understand the behavior of members and casual users based on data collected immediately after a trip. Both members and casual users completed the survey. The survey was implemented through QR codes posted at kiosks that brought the user to a survey link that they could take on their smartphone. The text of the URL was also provided, if respondents wanted to type it in or take it later. The application of the survey in its on-street application required that the user possess a smartphone. Three U.S. operators deployed the on-street survey. They included Hubway in Boston (N = 191), B-cycle in San Antonio (N = 14), and GREENBike SLC in Salt Lake City (N = 1). The distribution of membership from the three surveys is shown in Table 30.
Mineta Transportation Institute
88 Public Bikesharing User Survey Analysis
Table 30. On-Street Survey Respondent Distribution by Membership Type
Annual Member Monthly Member 7-Day Pass 3-Day Pass 24-Hour Pass Visiting Member Using B-connected Non- response Total N
Greenbike LLC 100% N/A 0% N/A 0% 0% 0% 1
San Antonio B-cycle 29% N/A 7% N/A 57% 0% 7% 14
Hubway 72% 4% N/A 4% 19% N/A 1% 191
Note: “N/A” denotes that a particular membership type was not offered by that operator.
The distribution of the surveys shows limited success in this experimental method for surveying casual users. In Salt Lake City, where the membership base was small at the time of the survey, only one valid respondent was collected via the on-street survey. Because GREENBike had a sample size of 1, the authors did not include it in the distributions that follow. San Antonio features a bigger system and had additional respondents. Finally, Hubway, which is the largest of the systems, had the greatest number of respondents. San Antonio had a majority of casual members responding to the survey (64%), whereas about 20% of Hubway respondents were casual members.
Respondents to the survey were asked a short number of questions related to their bikesharing use immediately following their trip. One of the first questions was related to trip purpose, which was asked of all respondents. The cross-tabulation of trip purpose by membership type is shown in Table 31 for both Hubway and San Antonio.
Mineta T
ransportation
Institute
89
Public Bikesharing User Survey
Analysis
Table 31. Cross-Tabulation of Membership Type by Trip Purpose
Trip Purpose Type of Membership Go to/from work Go to/from school Go to a meeting Go to a restaurant / meal Go shopping Social / entertainment / visit friends Run errands Exercise / recreation Other Hubway Annual Member 40% 3% 4% 3% 1% 6% 9% 4% 3%
Monthly Member 2% 0% 0% 0% 0% 1% 0% 2% 0%
3-Day Pass 0% 0% 2% 0% 1% 2% 0% 0% 1%
24-Hour Pass 2% 1% 2% 1% 1% 5% 2% 5% 4%
Total N 83 6 13 6 3 25 20 20 15
San Antonio Annual Member 0% 0% 0% 0% 0% 8% 0% 23% 0%
Monthly Member 0% 0% 0% 0% 0% 8% 0% 0% 0%
3-Day Pass 0% 8% 0% 0% 0% 0% 0% 38% 15%
24-Hour Pass 0% 0% 0% 0% 0% 0% 0% 0% 0%
Mineta Transportation Institute
90 Public Bikesharing User Survey Analysis
The cross-tabulation shows that members in Boston, which were predominantly Annual Members, used bikesharing for commuting to work, meetings, and other practical daily purposes. Fewer respondents in Boston listed social/recreational trips compared to San Antonio. In San Antonio, a sizable proportion of respondents who had a 3-day pass or an annual member pass used B-cycle for recreational purposes. This is not surprising given the focus of the San Antonio program on promoting public health.
Finally, the on-street survey asked a direct question about modal substitution, asking, if the user had not used bikesharing, how they would have made their most recent trip. The distribution of responses for Boston and San Antonio are shown in Figure 24. Please note that the one survey respondent from Salt Lake City stated he/she would not have made the trip in the absence of bikesharing. The distribution below supports findings from the member survey. Considerable substitution of bus and rail is observed in Boston, along with small shares of “Drive Alone” and “Taxi.”
San Antonio, although the sample size is far smaller, shows a more auto-centric substitution. This result, nevertheless, should be verified with a larger sample.
36% 14% 14% 21% 0% 7% 0% 7% 0% 0% 0% 5% 10% 15% 20% 25% 30% 35% 40% I would not have made this trip
Bus Personal bike Drive alone Drive with
others Ride in a carwith others Taxi Walk Carsharingvehicle Other (pleasespecify)
If bikesharing was not available, how would you have made this trip? (check the MAIN mode that you would have used)
San Antonio, N = 14 5% 15% 5% 3% 2% 0% 4% 31% 0% 2% 32% 2% 0% 0% 5% 10% 15% 20% 25% 30% 35% I would not have made this trip Bus Personal
bike Drive alone Drive withothers Ride in acar with
others
Taxi Walk Zipcar or
other carsharing vehicle Other (please specify) Subway or
trolley CommuterRail Ferry
Boston, N = 191
Figure 24. Modal Substitution of Public Bikesharing from On-Street Survey
In San Antonio, a sizable portion of respondents who had a 3-day pass or an annual member pass used B-cycle for recreational purposes.
Mineta Transportation Institute
91 Public Bikesharing User Survey Analysis
The on-street survey was an experimental survey method to “passive” recruitment, which aimed to achieve two objectives. First, it was designed to survey casual bikesharing users, which are not otherwise contactable. Second, it aimed to obtain information about trip purpose and substitution at or near the time of the trip. However, the on-street survey implementation was only marginally successful and encountered a few challenges. As an entirely passive survey, it lacked any human engagement and relied upon the attention and initiative of the respondent.
Although the QR was prominently displayed, the posting also had a short URL that respondents could use to access the survey, so a QR code was not required. Despite these limitations, a reasonable sample size was obtained in Boston, but it was less effective in producing a large sample (>30) in San Antonio or Salt Lake City. These lessons learned, as well as the data obtained from the existing sample, can be used to further improve research on casual users and to develop cost-effective methodological approaches for collecting data on this population.
Analysis of Activity and Survey Data
Collaboration with one public bikesharing operator (Nice Ride Minnesota) in this study permitted the anonymous linking of survey data to annual activity data for the year. The data provide unique possibilities for new analyses to understand how respondents use the bikesharing system in specific ways. The link occurred through a de-identified parameter, which contained no information about the respondent’s identity, but contained enough information to be matched to the survey and activity data of Nice Ride Minnesota. The cross-tabulation of the survey and activity data can help to verify that actual usage frequencies are at levels similar to those reported in the survey.
These data allowed researchers to analyze how modal shift correlates with use. For example, the researchers can investigate the distribution of modal shifts toward and away from driving and public transit at a regional level by combining the activity and survey data. An example of this kind of analysis is presented below. Figure 25 shows the average count of trips taken during a single year (2013) by respondents, as correlated with their “stated” modal shift in bus, rail, walking, and driving. Please note that the sample size for each mode is provided in the legend. The sample sizes are slightly smaller than the survey responses reported earlier due to missing observations (~15 to 20) in the activity data.
Mineta Transportation Institute
92 Public Bikesharing User Survey Analysis
0 20 40 60 80 100 120 140 160 180 Much more
often More often About thesame Less often Much lessoften Did not usebefore or now
Have changed use, not due
to bikesharing Av er ag e Tr ip s P er Y ea r
Walk, N = 603 Driving, N = 604 Bus, N = 604 Rail, N = 604
Figure 25. Average Number of Bikesharing Trips by Modal Shift
The data in Figure 25 represent the average number of annual bikesharing trips of respondents by modal shift. It shows that those shifting away from all modes tended to use bikesharing with greater frequency. This makes sense, as frequent bikesharing users find the system attractive and substitute their previous travel modes with higher bikesharing use. The data also suggest that those shifting toward certain modes are somewhat regular users, using bikesharing between an average of 42 to 71 times per year. These connections may be useful in yielding new understanding about the dynamics of bikesharing impacts, activity, and behavior in the future.
Mineta Transportation Institute
93
VIII. CONCLUSION AND LESSONS LEARNED
Public bikesharing systems offers users access to bicycles on an as-needed basis for first- and-last mile trips connecting to other modes, as well as for both short- and long-distance destinations in an urban environment. Between 2007 and December 2013, there were 44 IT-based public bikesharing startups, three program suspensions, and three program closures in North America. A number of public bikesharing business models have evolved in North America with the advent of IT-based systems including: 1) non-profit, 2) privately owned and operated, 3) publicly owned and operated, 4) public owned/contractor operated, and 5) vendor operated. In the 2012 season, there were 28 IT-based public bikesharing programs with approximately 1.1 million users sharing 17,344 bicycles at 1,599 locations in North America. In North America, casual (short-term) users accounted for 85% of all bikesharing users during the 2012 season. Globally, as of June 2014, public bikesharing programs existed on five continents, including 712 cities, operating approximately 806,200 bicycles at 37,500 stations (Russell Meddin, unpublished data, June 2014).
This study examined public bikesharing from several angles including: 1) current operational practices, 2) business models, 3) membership demographics, and 4) environmental and social impacts in North America. A combined 70 interviews were conducted with local government representatives and operators during Phase I and Phase II of this study. In addition to expert interviews, five operators participated in a member survey with 6,373 individual responses, and three operators participated in a survey of casual users with 205 individual respones. The recent proliferation of IT-based public bikesharing operations have led to a range of critical observations and lessons learned. This study revealed the following key findings as summarized below.
Bicycle Theft and Vandalism
Early bikesharing programs learned that user anonymity created systems prone to theft. The world’s first documented bikesharing program, Amsterdam’s Witte Fietsenplan, commonly referred to as “White Bikes,” saw the majority of the system’s bikes disappear just days after its launch in Summer 1965. Many of the bikes, which were left unlocked for anyone to use, were either confiscated by the police, stolen, or thrown into local bodies of water. Other first-generation systems, such as the “Yellow Bike Project” of Portland in 1994 and the “Purple People Movers” of Phoenix in 1997, succumbed to similar fates as the majority of bikes were stolen within a few months of each system’s launch.
To address this issue of user anonymity, the next generation of bikesharing programs employed technology that required users to supply a small deposit that would be returned to the user once the bicycle was returned (e.g., second generation systems). While some second-generation systems are still in use, the deposit values are generally significantly less than the value of the bicycle; therefore, theft and vandalism remain a prevalent issue in second-generation systems. To address this, IT-bikesharing systems introduced electronic smartcards to record user information, access bicycles, and track usage. These features, which mark the third generation of bikesharing equipment, maintain accountability as users can face fines of at least US$1,000 for bicycles lost while in their possession. Vandalism and theft are both reported to be very low in systems featuring third-generation technology.
Mineta Transportation Institute
94 Conclusion and Lessons Learned
Bicycle Redistribution
Bicycle redistribution or “rebalancing” remains a daily challenge for operators. In any bikesharing program, the operator must be able to effectively manage bikesharing bicycles and available docks to prevent scenarios in which a station is entirely full or empty. Rebalancing requires real-time information on the location of bicycles and operational equipment and labor to rebalance bicycles. Using natural gas-powered vans, trucks, or trailers, rebalancers move bicycles from areas of high-bicycle density to areas of low- bicycle density, depending on daily usage patterns and forecasts.
In some cases, rebalancing requirements are written into contracts between the relevant governmental agency and the program operator, requiring that a full or empty station not remain as such beyond a certain time period. It also remains a primary customer service issue and one of the highest costs relating to operations, especially for programs with high bicycle use.
Helmet Considerations
Helmet laws also present a concern for bikesharing programs. Compulsory all-age helmet laws have been reported to restrict a bikesharing system’s use as a vast majority of bikesharing users report “rarely” or “never” wear a helmet while using bikesharing. Furthermore, reported bikesharing crash statistics show that bikesharing may be safer than regular cycling, as reported in Chapter 5, “Public Bikesharing Operations.” This is likely attributed to the considerable weight of the bicycle, the presence of reflectors and lights, the bicycle’s low center of gravity, and the gear ratio, which generally prevents cyclists from riding at high speeds.
The City of Dallas, Texas, which is currently planning a bikesharing program, recently revoked its compulsory helmet law in light of the poor performance of bikesharing programs subject to such regulations. In other cities with compulsory laws, such as Seattle and Vancouver, further development of helmet dispensing and sanitizing systems could increase helmet usage and possibly the number of bikesharing participants.
Role of Supportive Infrastructure and Partnerships
Many program operators have cited that establishing partnerships within local government and with community stakeholders is imperative to successful bikesharing operations. Prior to a system’s launch, operators should work with relevant city agencies and staff to improve bicycle infrastructure that will be necessary to support the increase of cyclists generated by the bikesharing program.
An operator should have a keen understanding of city policies and agencies so that prospective hurdles in planning and implementation can be addressed effectively. Additionally, establishing relationships with local cycling advocates is also imperative in generating strong support for the bikesharing program. Other relevant community groups should also be contacted for purposes of outreach, system planning, and marketing the program to the local population.
Mineta Transportation Institute
95 Conclusion and Lessons Learned
Prelaunch Considerations
Successful public bikesharing programs are those that address the specific needs of their users and market segments prior to and after deployment. Appropriate spatial analysis to properly locate bikesharing stations is imperative to system use, in addition to employing station technology that is mobile and can be relocated according to usage patterns. Operators should also allow for proper public engagement through both public forums and Internet “suggest-a-station” platforms.
Additionally, cities can alter cycling infrastructure and policies prior to a bikesharing program launch, including those that require all-age helmet use. Furthermore, prelaunch marketing and general outreach is critical for success.
Different Users Account for Different Usage and Revenue
While having a strong foundation of annual members is important to a system’s success, tailoring components of the system to encourage use by the casual user is imperative for a system’s long-term economic viability, especially in lieu of public subsidy. This finding has been further emphasized by recent developments relating to New York City’s Citi Bike program and its apparent revenue shortfall. At present, Citi Bike has a considerably lower proportion of casual to annual users in contrast to cities, such as Washington, D.C.
The Need for Social Equity Planning, Incentives and Marketing
Data have shown that bikesharing users are more likely be male, Caucasian, wealthier,